Even in the absence of cardiovascular disease, persons with NIDDM have an impaired ability to carry out maximal exercise. In addition, oxygen consumption (VO2) at submaximal work loads is reduced during graded exercise in NIDDM compared to healthy controls. The lower VO2 per work load observed in NIDDM has previously been observed in disease states where O2 delivery is impaired and VO2 kinetics are slowed. Finally, slowed oxygen kinetics and heart rate kinetics during constant load exercise also occur in NIDDM. The functional significance of a decreased exercise performance is that there is a direct correlation between VO2 max and habitual physical activity in persons with NIDDM. Importantly, decreased habitual physical activity is known to be associated with poorer diabetic status. In addition, it has been shown in a recent population study that there was a correlation between all cause mortality and low levels of fitness (measured using a testing protocol on which exercise performance correlates highly with VO2 max). Therefore, improvement in exercise tolerance in NIDDM could have multiple benefits for patients in terms of improving functional capacity and reducing morbidity. Pilot data suggests that the cardiovascular exercise impairments observed may directly relate to a decreased ability to increase cardiac output and/or arteriovenous oxygen difference with abnormalities have not been identified. No correlation has been found between exercise performance and aspects of glucose metabolism, such as HbA1C, fasting serum glucose or insulin concentrations. However, the association of increased insulin sensitivity and exercise performance has not been well investigated. This relationship is of interest since the insulin sensitivity in diabetic animals as well as in humans with diabetes. The mechanisms accounting for this effect and their tissue sensitivity, could potentiate the cardiovascular aspects of insulin action (i.e. improve exercise performance) either by improving cardiac output, blood flow distribution or both.